Dr

Background:  Infant siblings of children with autism spectrum disorder (at-risk sibs) –who are themselves at genetic risk of autism –are more likely to exhibit early social communicative impairments, and other atypicalities, than typically developing siblings (low-risk sibs). Taking a developmental transactional model, the appearance of early infant atypicalities in those infants at risk may disrupt parent-infant interaction which may in turn amplify existing social atypicalities through their experiencing of less optimal social interactive environments. This theory does not suggest that parent-infant interactions are a primary cause of atypical behaviour. A few studies to date suggest that early parent-infant interactions show specific subtle but consistent differences in at-risk sibs in early-middle infancy. The current study follows up a group of at-risk sibs for whom group differences were found in parent non-directiveness/sensitive responsiveness and infant liveliness at 6-10 months during parent-infant interaction.

Objectives:  (1) To compare the global mother-infant play interaction characteristics between at-risk sib infants at 12-15 months with low-risk sib controls; (2) To examine whether 12-month parent-infant interaction predicts 24-month ADOS score beyond the contribution of early behavioural atypicalities (AOSI score).

Methods:  Forty-four at-risk and 48 low-risk sibs were videotaped in 6-min mother-infant unstructured play interactions within the British Autism Study for Infant Siblings (BASIS) rated, blind to dyad information, on a global rating scheme which involved 2 parental, 3 infant and 2 dyadic scales. The Autism Observation Scale for Infants (AOSI) was administered independent of interaction coding at 12 months, and the at-risk sib group were administered the Autism Diagnostic Observation Schedule (ADOS) at 24 months.

Results:  Compared to the low-risk group, the at-risk group showed significantly lower scores in parent sensitive responsiveness, parent non-directiveness, infant attentiveness to parent, and mutuality, after controlling for infant age. Twelve-month AOSI score predicted 24-month symptomatology. The four areas of interaction, tested in individual linear regression models, predicted 24-month ADOS score independent of 12-month AOSI and infant age/developmental level, with dyadic mutuality as a particularly strong predictor. In all models, the variance accounted for by AOSI score was no longer significant, except that which tested parent non-directiveness, in which both were independent predictors. 

Conclusions:  Parent, infant and dyadic features of parent-infant interaction, which were identified as being associated with ASD risk at 12 months, were stronger predictors of 24-month ADOS score than 12-month markers (AOSI), lending support to the transactional model. The findings support the role of parent-infant interaction in emergent ASD development – along the lines being targeted in by very early (prodromal) intervention, such as in iBASIS.